Class DistanceTwoLinkPredictionIndividualSamplerConfigurator<U>

java.lang.Object
es.uam.eps.ir.relison.grid.links.recommendation.sampling.DistanceTwoLinkPredictionIndividualSamplerConfigurator<U>
Type Parameters:
U - type of the users.
All Implemented Interfaces:
IndividualSamplingAlgorithmConfigurator<U>

public class DistanceTwoLinkPredictionIndividualSamplerConfigurator<U>
extends java.lang.Object
implements IndividualSamplingAlgorithmConfigurator<U>
Class for selecting nodes at distance 2 from the target user, considering the usual approach in link prediction, i.e. we first take all the nodes at distance 2 from the target user in the training set. Then, we take all nodes in that set with whom the target user has created a link in the test set, and we randomly select the same number of negative examples.
  • Field Summary

    Fields 
    Modifier and Type Field Description
    private static java.lang.String USEL
    Identifier of the neighbor orientation for the sampled user.
    private static java.lang.String VSEL
    Identifier of the neighbor orientation for the objective users.
  • Constructor Summary

    Constructors 
    Constructor Description
    DistanceTwoLinkPredictionIndividualSamplerConfigurator()  
  • Method Summary

    Modifier and Type Method Description
    org.jooq.lambda.tuple.Tuple2<java.lang.String,​IndividualSamplerFunction<U>> grid​(Parameters params)
    Obtains a function for obtaining an individual sampling algorithm.
    org.jooq.lambda.tuple.Tuple2<java.lang.String,​IndividualSampler<U>> grid​(Parameters params, FastGraph<U> trainGraph, FastGraph<U> testGraph, es.uam.eps.ir.ranksys.fast.preference.FastPreferenceData<U,​U> prefData)
    Obtains an individual sampling algorithm.

    Methods inherited from class java.lang.Object

    clone, equals, finalize, getClass, hashCode, notify, notifyAll, toString, wait, wait, wait